ABSTRACT
Neural network training can be slow and energy-expensive due to the frequent transfer of weight data between digital memory and processing units. Neuromorphic systems can accelerate neural networks by performing multiply-accumulate operations in parallel using nonvolatile analog memory. However, executing the widely used backpropagation training algorithm in multilayer neural networks requires information-and therefore storage-of the partial derivatives of the weight values preventing suitable and scalable implementation in hardware. Here, we propose a hardware implementation of the backpropagation algorithm that progressively updates each layer using in situ stochastic gradient descent, avoiding this storage requirement. We experimentally demonstrate the in situ error calculation and the proposed progressive backpropagation method in a multilayer hardware-implemented neural network. We confirm identical learning characteristics and classification performance compared to conventional backpropagation in software. We show that our approach can be scaled to large and deep neural networks, enabling highly efficient training of advanced artificial intelligence computing systems.
ABSTRACT
Accurate glucose prediction is vital for diabetes management. Artificial intelligence and artificial neural networks (ANNs) are showing promising results for reliable glucose predictions, offering timely warnings for glucose fluctuations. The translation of these software-based ANNs into dedicated computing hardware opens a route toward automated insulin delivery systems ultimately enhancing the quality of life for diabetic patients. ANNs are transforming this field, potentially leading to implantable smart prediction devices and ultimately to a fully artificial pancreas. However, this transition presents several challenges, including the need for specialized, compact, lightweight, and low-power hardware. Organic polymer-based electronics are a promising solution as they have the ability to implement the behavior of neural networks, operate at low voltage, and possess key attributes like flexibility, stretchability, and biocompatibility. Here, the study focuses on implementing software-based neural networks for glucose prediction into hardware systems. How to minimize network requirements, downscale the architecture, and integrate the neural network with electrochemical neuromorphic organic devices, meeting the strict demands of smart implants for in-body computation of glucose prediction is investigated.
Subject(s)
Blood Glucose , Neural Networks, Computer , Humans , Artificial Intelligence , Blood Glucose/analysis , SoftwareABSTRACT
Fuzzy logic is useful tool to describe and represent biological or medical scenarios, where often states and outcomes are not only completely true or completely false, but rather partially true or partially false. Despite its usefulness and spread, fuzzy logic modeling might easily be done in the wrong way, especially by beginners and unexperienced researchers, who might overlook some important aspects or might make common mistakes. Malpractices and pitfalls, in turn, can lead to wrong or overoptimistic, inflated results, with negative consequences to the biomedical research community trying to comprehend a particular phenomenon, or even to patients suffering from the investigated disease. To avoid common mistakes, we present here a list of quick tips for fuzzy logic modeling any biomedical scenario: some guidelines which should be taken into account by any fuzzy logic practitioner, including experts. We believe our best practices can have a strong impact in the scientific community, allowing researchers who follow them to obtain better, more reliable results and outcomes in biomedical contexts.
Subject(s)
Fuzzy Logic , Medicine , HumansABSTRACT
PURPOSE: Congenital myopathies are a heterogeneous group of diseases affecting the skeletal muscles and characterized by high clinical, genetic, and histological variability. Magnetic Resonance (MR) is a valuable tool for the assessment of involved muscles (i.e., fatty replacement and oedema) and disease progression. Machine Learning is becoming increasingly applied for diagnostic purposes, but to our knowledge, Self-Organizing Maps (SOMs) have never been used for the identification of the patterns in these diseases. The aim of this study is to evaluate if SOMs may discriminate between muscles with fatty replacement (S), oedema (E) or neither (N). METHODS: MR studies of a family affected by tubular aggregates myopathy (TAM) with the histologically proven autosomal dominant mutation of the STIM1 gene, were examined: for each patient, in two MR assessments (i.e., t0 and t1, the latter after 5 years), fifty-three muscles were evaluated for muscular fatty replacement on the T1w images, and for oedema on the STIR images, for reference. Sixty radiomic features were collected from each muscle at t0 and t1 MR assessment using 3DSlicer software, in order to obtain data from images. A SOM was created to analyze all datasets using three clusters (i.e., 0, 1 and 2) and results were compared with radiological evaluation. RESULTS: Six patients with TAM STIM1-mutation were included. At t0 MR assessments, all patients showed widespread fatty replacement that intensifies at t1, while oedema mainly affected the muscles of the legs and appears stable at follow-up. All muscles with oedema showed fatty replacement, too. At t0 SOM grid clustering shows almost all N muscles in Cluster 0 and most of the E muscles in Cluster 1; at t1 almost all E muscles appear in Cluster 1. CONCLUSION: Our unsupervised learning model appears to be able to recognize muscles altered by the presence of edema and fatty replacement.
Subject(s)
Myopathies, Structural, Congenital , Unsupervised Machine Learning , Humans , Muscle, Skeletal/diagnostic imaging , Muscle, Skeletal/pathology , Magnetic Resonance Imaging/methods , Myopathies, Structural, Congenital/diagnostic imaging , Myopathies, Structural, Congenital/genetics , Magnetic Resonance Spectroscopy , Edema/diagnostic imaging , Edema/pathology , Stromal Interaction Molecule 1/genetics , Neoplasm ProteinsABSTRACT
Introduction: Oxford Nanopore Technologies (ONT) is a third generation sequencing approach that allows the analysis of individual, full-length nucleic acids. ONT records the alterations of an ionic current flowing across a nano-scaled pore while a DNA or RNA strand is threading through the pore. Basecalling methods are then leveraged to translate the recorded signal back to the nucleic acid sequence. However, basecall generally introduces errors that hinder the process of barcode demultiplexing, a pivotal task in single-cell RNA sequencing that allows for separating the sequenced transcripts on the basis of their cell of origin. Methods: To solve this issue, we present a novel framework, called UNPLEX, designed to tackle the barcode demultiplexing problem by operating directly on the recorded signals. UNPLEX combines two unsupervised machine learning methods: autoencoders and self-organizing maps (SOM). The autoencoders extract compact, latent representations of the recorded signals that are then clustered by the SOM. Results and Discussion: Our results, obtained on two datasets composed of in silico generated ONT-like signals, show that UNPLEX represents a promising starting point for the development of effective tools to cluster the signals corresponding to the same cell.
ABSTRACT
Calcium homeostasis and signaling processes in Saccharomyces cerevisiae, as well as in any eukaryotic organism, depend on various transporters and channels located on both the plasma and intracellular membranes. The activity of these proteins is regulated by a number of feedback mechanisms that act through the calmodulin-calcineurin pathway. When exposed to hypotonic shock (HTS), yeast cells respond with an increased cytosolic calcium transient, which seems to be conditioned by the opening of stretch-activated channels. To better understand the role of each channel and transporter involved in the generation and recovery of the calcium transient-and of their feedback regulations-we defined and analyzed a mathematical model of the calcium signaling response to HTS in yeast cells. The model was validated by comparing the simulation outcomes with calcium concentration variations before and during the HTS response, which were observed experimentally in both wild-type and mutant strains. Our results show that calcium normally enters the cell through the High Affinity Calcium influx System and mechanosensitive channels. The increase of the plasma membrane tension, caused by HTS, boosts the opening probability of mechanosensitive channels. This event causes a sudden calcium pulse that is rapidly dissipated by the activity of the vacuolar transporter Pmc1. According to model simulations, the role of another vacuolar transporter, Vcx1, is instead marginal, unless calcineurin is inhibited or removed. Our results also suggest that the mechanosensitive channels are subject to a calcium-dependent feedback inhibition, possibly involving calmodulin. Noteworthy, the model predictions are in accordance with literature results concerning some aspects of calcium homeostasis and signaling that were not specifically addressed within the model itself, suggesting that it actually depicts all the main cellular components and interactions that constitute the HTS calcium pathway, and thus can correctly reproduce the shaping of the calcium signature by calmodulin- and calcineurin-dependent complex regulations. The model predictions also allowed to provide an interpretation of different regulatory schemes involved in calcium handling in both wild-type and mutants yeast strains. The model could be easily extended to represent different calcium signals in other eukaryotic cells.
ABSTRACT
Mathematical models of biochemical networks can largely facilitate the comprehension of the mechanisms at the basis of cellular processes, as well as the formulation of hypotheses that can be tested by means of targeted laboratory experiments. However, two issues might hamper the achievement of fruitful outcomes. On the one hand, detailed mechanistic models can involve hundreds or thousands of molecular species and their intermediate complexes, as well as hundreds or thousands of chemical reactions, a situation generally occurring in rule-based modeling. On the other hand, the computational analysis of a model typically requires the execution of a large number of simulations for its calibration, or to test the effect of perturbations. As a consequence, the computational capabilities of modern Central Processing Units can be easily overtaken, possibly making the modeling of biochemical networks a worthless or ineffective effort. To the aim of overcoming the limitations of the current state-of-the-art simulation approaches, we present in this paper FiCoS, a novel "black-box" deterministic simulator that effectively realizes both a fine-grained and a coarse-grained parallelization on Graphics Processing Units. In particular, FiCoS exploits two different integration methods, namely, the Dormand-Prince and the Radau IIA, to efficiently solve both non-stiff and stiff systems of coupled Ordinary Differential Equations. We tested the performance of FiCoS against different deterministic simulators, by considering models of increasing size and by running analyses with increasing computational demands. FiCoS was able to dramatically speedup the computations up to 855×, showing to be a promising solution for the simulation and analysis of large-scale models of complex biological processes.
Subject(s)
Biochemical Phenomena , Software , Systems Biology , Algorithms , Autophagy , Computational Biology , Computer Graphics , Computer Simulation , Humans , Mathematical Concepts , Metabolic Networks and Pathways , Models, Biological , Protein Biosynthesis , Synthetic BiologyABSTRACT
Combination therapies proved to be a valuable strategy in the fight against cancer, thanks to their increased efficacy in inducing tumor cell death and in reducing tumor growth, metastatic potential, and the risk of developing drug resistance. The identification of effective combinations of drug targets generally relies on costly and time consuming processes based on in vitro experiments. Here, we present a novel computational approach that, by integrating dynamic fuzzy modeling with multi-objective optimization, allows to efficiently identify novel combination cancer therapies, with a relevant saving in working time and costs. We tested this approach on a model of oncogenic K-ras cancer cells characterized by a marked Warburg effect. The computational approach was validated by its capability in finding out therapies already known in the literature for this type of cancer cell. More importantly, our results show that this method can suggest potential therapies consisting in a small number of molecular targets. In the model of oncogenic K-ras cancer cells, for instance, we identified combination of up to three targets, which affect different cellular pathways that are crucial for cancer proliferation and survival.
ABSTRACT
Surfing in rough waters is not always as fun as wave riding the "big one". Similarly, in optimization problems, fitness landscapes with a huge number of local optima make the search for the global optimum a hard and generally annoying game. Computational Intelligence optimization metaheuristics use a set of individuals that "surf" across the fitness landscape, sharing and exploiting pieces of information about local fitness values in a joint effort to find out the global optimum. In this context, we designed surF, a novel surrogate modeling technique that leverages the discrete Fourier transform to generate a smoother, and possibly easier to explore, fitness landscape. The rationale behind this idea is that filtering out the high frequencies of the fitness function and keeping only its partial information (i.e., the low frequencies) can actually be beneficial in the optimization process. We prove our theory by combining surF with a settings free variant of Particle Swarm Optimization (PSO) based on Fuzzy Logic, called Fuzzy Self-Tuning PSO. Specifically, we introduce a new algorithm, named F3ST-PSO, which performs a preliminary exploration on the surrogate model followed by a second optimization using the actual fitness function. We show that F3ST-PSO can lead to improved performances, notably using the same budget of fitness evaluations.
ABSTRACT
The investigation of cell proliferation can provide useful insights for the comprehension of cancer progression, resistance to chemotherapy and relapse. To this aim, computational methods and experimental measurements based on in vivo label-retaining assays can be coupled to explore the dynamic behavior of tumoral cells. ProCell is a software that exploits flow cytometry data to model and simulate the kinetics of fluorescence loss that is due to stochastic events of cell division. Since the rate of cell division is not known, ProCell embeds a calibration process that might require thousands of stochastic simulations to properly infer the parameterization of cell proliferation models. To mitigate the high computational costs, in this paper we introduce a parallel implementation of ProCell's simulation algorithm, named cuProCell, which leverages Graphics Processing Units (GPUs). Dynamic Parallelism was used to efficiently manage the cell duplication events, in a radically different way with respect to common computing architectures. We present the advantages of cuProCell for the analysis of different models of cell proliferation in Acute Myeloid Leukemia (AML), using data collected from the spleen of human xenografts in mice. We show that, by exploiting GPUs, our method is able to not only automatically infer the models' parameterization, but it is also 237× faster than the sequential implementation. This study highlights the presence of a relevant percentage of quiescent and potentially chemoresistant cells in AML in vivo, and suggests that maintaining a dynamic equilibrium among the different proliferating cell populations might play an important role in disease progression.
Subject(s)
Algorithms , Computer Graphics , Animals , Cell Proliferation , Computer Simulation , Flow Cytometry , Mice , SoftwareABSTRACT
MOTIVATION: The elucidation of dysfunctional cellular processes that can induce the onset of a disease is a challenging issue from both the experimental and computational perspectives. Here we introduce a novel computational method based on the coupling between fuzzy logic modeling and a global optimization algorithm, whose aims are to (1) predict the emergent dynamical behaviors of highly heterogeneous systems in unperturbed and perturbed conditions, regardless of the availability of quantitative parameters, and (2) determine a minimal set of system components whose perturbation can lead to a desired system response, therefore facilitating the design of a more appropriate experimental strategy. RESULTS: We applied this method to investigate what drives K-ras-induced cancer cells, displaying the typical Warburg effect, to death or survival upon progressive glucose depletion. The optimization analysis allowed to identify new combinations of stimuli that maximize pro-apoptotic processes. Namely, our results provide different evidences of an important protective role for protein kinase A in cancer cells under several cellular stress conditions mimicking tumor behavior. The predictive power of this method could facilitate the assessment of the response of other complex heterogeneous systems to drugs or mutations in fields as medicine and pharmacology, therefore paving the way for the development of novel therapeutic treatments. AVAILABILITY AND IMPLEMENTATION: The source code of FUMOSO is available under the GPL 2.0 license on GitHub at the following URL: https://github.com/aresio/FUMOSO. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
Subject(s)
Neoplasms , Software , Algorithms , Humans , MutationABSTRACT
Advances in microscopy imaging technologies have enabled the visualization of live-cell dynamic processes using time-lapse microscopy imaging. However, modern methods exhibit several limitations related to the training phases and to time constraints, hindering their application in the laboratory practice. In this work, we present a novel method, named Automated Cell Detection and Counting (ACDC), designed for activity detection of fluorescent labeled cell nuclei in time-lapse microscopy. ACDC overcomes the limitations of the literature methods, by first applying bilateral filtering on the original image to smooth the input cell images while preserving edge sharpness, and then by exploiting the watershed transform and morphological filtering. Moreover, ACDC represents a feasible solution for the laboratory practice, as it can leverage multi-core architectures in computer clusters to efficiently handle large-scale imaging datasets. Indeed, our Parent-Workers implementation of ACDC allows to obtain up to a 3.7× speed-up compared to the sequential counterpart. ACDC was tested on two distinct cell imaging datasets to assess its accuracy and effectiveness on images with different characteristics. We achieved an accurate cell-count and nuclei segmentation without relying on large-scale annotated datasets, a result confirmed by the average Dice Similarity Coefficients of 76.84 and 88.64 and the Pearson coefficients of 0.99 and 0.96, calculated against the manual cell counting, on the two tested datasets.
ABSTRACT
BACKGROUND: In order to fully characterize the genome of an individual, the reconstruction of the two distinct copies of each chromosome, called haplotypes, is essential. The computational problem of inferring the full haplotype of a cell starting from read sequencing data is known as haplotype assembly, and consists in assigning all heterozygous Single Nucleotide Polymorphisms (SNPs) to exactly one of the two chromosomes. Indeed, the knowledge of complete haplotypes is generally more informative than analyzing single SNPs and plays a fundamental role in many medical applications. RESULTS: To reconstruct the two haplotypes, we addressed the weighted Minimum Error Correction (wMEC) problem, which is a successful approach for haplotype assembly. This NP-hard problem consists in computing the two haplotypes that partition the sequencing reads into two disjoint sub-sets, with the least number of corrections to the SNP values. To this aim, we propose here GenHap, a novel computational method for haplotype assembly based on Genetic Algorithms, yielding optimal solutions by means of a global search process. In order to evaluate the effectiveness of our approach, we run GenHap on two synthetic (yet realistic) datasets, based on the Roche/454 and PacBio RS II sequencing technologies. We compared the performance of GenHap against HapCol, an efficient state-of-the-art algorithm for haplotype phasing. Our results show that GenHap always obtains high accuracy solutions (in terms of haplotype error rate), and is up to 4× faster than HapCol in the case of Roche/454 instances and up to 20× faster when compared on the PacBio RS II dataset. Finally, we assessed the performance of GenHap on two different real datasets. CONCLUSIONS: Future-generation sequencing technologies, producing longer reads with higher coverage, can highly benefit from GenHap, thanks to its capability of efficiently solving large instances of the haplotype assembly problem. Moreover, the optimization approach proposed in GenHap can be extended to the study of allele-specific genomic features, such as expression, methylation and chromatin conformation, by exploiting multi-objective optimization techniques. The source code and the full documentation are available at the following GitHub repository: https://github.com/andrea-tango/GenHap .
Subject(s)
Algorithms , Computational Biology/methods , Haplotypes/genetics , Databases, Genetic , Humans , Time FactorsABSTRACT
MOTIVATION: Acute myeloid leukemia (AML) is one of the most common hematological malignancies, characterized by high relapse and mortality rates. The inherent intra-tumor heterogeneity in AML is thought to play an important role in disease recurrence and resistance to chemotherapy. Although experimental protocols for cell proliferation studies are well established and widespread, they are not easily applicable to in vivo contexts, and the analysis of related time-series data is often complex to achieve. To overcome these limitations, model-driven approaches can be exploited to investigate different aspects of cell population dynamics. RESULTS: In this work, we present ProCell, a novel modeling and simulation framework to investigate cell proliferation dynamics that, differently from other approaches, takes into account the inherent stochasticity of cell division events. We apply ProCell to compare different models of cell proliferation in AML, notably leveraging experimental data derived from human xenografts in mice. ProCell is coupled with Fuzzy Self-Tuning Particle Swarm Optimization, a swarm-intelligence settings-free algorithm used to automatically infer the models parameterizations. Our results provide new insights on the intricate organization of AML cells with highly heterogeneous proliferative potential, highlighting the important role played by quiescent cells and proliferating cells characterized by different rates of division in the progression and evolution of the disease, thus hinting at the necessity to further characterize tumor cell subpopulations. AVAILABILITY AND IMPLEMENTATION: The source code of ProCell and the experimental data used in this work are available under the GPL 2.0 license on GITHUB at the following URL: https://github.com/aresio/ProCell. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.